UC Berkeley

This dissertation is composed of three chapters and studies issues related to crime, unemployment and health. The first chapter looks at the effect of funding for public safety on drug related violence. The second chapter, which is joint work with Santiago Guerrero, examines the effect of unemployment on crime during the latest great recession. The third chapter examines the effect that the introduction of ultra-low sulfur diesel has had on infant mortality.

The first paper develops a simple framework to describe the effect of increases in fighting capacity on violence and uses a large program in Mexico to empirically estimate the effect of funding for public safety on violence, specifically drug related violence. Starting in 2008, Mexico implemented a large program designed for the strengthening of the municipal police, the assignment of which was based on an index. The main areas of allowed expenditures for these funds were: the purchase of fighting equipment, technology infrastructure and training of the police force. Instrumenting funding with the arbitrary initial eligibility cutoff, I find that the funds led to large increases in drug related violence. Evidence is consistent with the funds allowing the police to fight criminal organizations which weakened organizations and in turn led to turf wars. The effect is not higher for PAN municipalities, the party whose main platform during the study period was to fight organized crime. Also, there does not seem to be an increase in violence in politically stable municipalities as a result of the program, but there is a decrease in areas with low land productivity. Consistent with theory, I also find suggestive evidence of an inverted U-shaped relationship between baseline funding for public safety and the effect of the program.

The second paper estimates the effect of unemployment on crime in Mexico. This study uses the variation in unemployment across metropolitan areas in Mexico induced by the latest great recession. Areas that were highly dependent on the US economy experienced the largest increases in unemployment, thus we instrument unemployment with the initial manufacturing and tourism labor share interacted with US GDP and find that increases in unemployment have led to decreases in crime in Mexico. The results are consistent with the decrease in potential targets due to the increases in unemployment outweighing the positive effect coming from the decrease in the opportunity cost of engaging in criminal activities as unemployment increases.

The third paper estimates the effect of the introduction of ultra-low sulfur diesel on infant health in Mexico. In 2006 the Mexican government began the rollout of ultra-low sulfur diesel in metropolitan areas, starting with border municipalities. Using a difference in differences approach, I find that, despite its potential to improve health outcomes, there is no evidence that sulfur regulation had a substantial effect on infant mortality outcomes.